| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online May 8, 2003 as doi:10.1096/fj.02-0980fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Taub Institute for Research on Alzheimer's Disease and the Aging Brain, Department of Pathology, College of Physicians and Surgeons, Columbia University, P&S 15-408, New York, New York, USA; and
* Department of Virology and
Laboratory of Digital Microscopy, NYS Institute for Basic Research, Staten Island, New York, USA
2Correspondence: Taub Institute for Research on Alzheimer's Disease and the Aging Brain, Department of Pathology, College of Physicians and Surgeons, Columbia University, P&S 15-408, 630 W. 168th St., New York, NY 10032, USA. E-mail: nt2024{at}columbia.edu
SPECIFIC AIM
We aimed to explore further the physiological significance of the interaction between PS1 and CLIP-170 and investigate the consequences of its disruption. We addressed the relevance of this interaction to pathways leading to production and clearance of Aß. Our approach involved mapping of their binding domains (BDPs) and the transfection of cells with vectors containing these domains. This was followed by monitoring the formation of the PS1/CLIP-170 complex, consisting of the corresponding full-length proteins, along with the measurement of the levels of Aß formed and secreted in the culture media and the amount of exogenous Aß taken up by the cells.
PRINCIPAL FINDINGS
1. Transfection of SY5Y cells with vectors driving the expression of the 600 amino acid carboxyl-terminal region of CLIP-170 or the 150 amino acid large cytoplasmic loop of PS1 disrupt the formation of the PS1/CLIP-170 complex
We generated SY5Y cell-lines stably transfected with the pCI-neo vector (Promega, Madison, WI, USA) modified to express full-length PS1 or the previously determined BDPs (fragments PS1251-458 or CLIP-170792-1392). Selected clones with high levels of expression of the BDPs (see Fig. 1C
for PS1L) were then transiently transfected with Myc-CLIP-170. Figure 1A
shows that expression of either BDP prevented the formation of the complex between full-length PS1 and CLIP-170 in cultured cells.
|
2. Mapping of the minimum linear peptide fragments of PS1 and CLIP-170 required for their binding
As smaller peptides are more amenable to pharmacological manipulation, we mapped the minimum binding domains (mBDPs) of PS1 and CLIP-170. This was accomplished by screening 
gt11 expression minilibraries by in vitro binding assays, using purified GST-fusion peptides corresponding to the binding domains. The minilibraries were obtained from the original clones containing the BDPs by restriction enzyme digestions and religation into the bacteriophage DNA. The procedure identified the 60 amino acid amino-terminal portion of the large cytoplasmic domain of PS1 and the 30 amino acid carboxyl-terminal end of CLIP-170 as mBDPs. BDPs and mBDPs were examined for their effectiveness in modulating the formation of the PS1/CLIP-170 complex in addition to Aß formation and its uptake from the medium.
3. Disruption of the PS1/CLIP-170 complex is associated with a lower level of Aß in the culture media
The lower amount of the PS1/CLIP-170 complex correlated with lower levels of newly synthesized Aß as detected by 35S-Met metabolic labeling, followed by its immunoprecipitation from the media (Fig. 1B
, lanes 3, 4). There was a significant negative correlation (regression analysis: multiple R=–0.969, P=0.0001, n=3, in triplicate) between the amount of PS1L fragment expressed by a given transfected clone and the amount of Aß detected in the media (Fig. 1C
).
To verify whether the mBDPs would have a similar effect, we transiently transfected SY5Y cells with plasmids engineered to contain the following mBDP-expressing cDNA inserts: CLIP-1701363-1392, CLIP-1701183-1392, PS1251-312, PS1251-284 (weakly positive); and the non-BDP-expressing cDNA inserts CLIP-1701183-1362, CLIP-170789-873, PS1313-397, PS1416-458 (numbering refers to the amino acid position of the corresponding protein). Each DNA fragment was excised from the original phage DNA, cloned into the EcoRI site of the pCI-neo vector, then transiently transfected in SY5Y cells. Only BDPs as determined by the in vitro screening procedure described above were capable of lowering Aß (white bars, Fig. 1D
). The PS1251-281 peptide, which bound weakly to CLIP-170, was an inefficient modulator of Aß formation. These results further established a functional link between the interaction of PS1 with CLIP-170 and Aß formation.
The expression of the BDPs or smaller fragments thereof in SY5Y cells had no effect on the levels of full-length APP as examined by pulse-chase experiments (not shown). In contrast, examination of the carboxyl-terminal proteolytic fragments of APP showed that the decreased levels of Aß caused by the expression of BDPs (PS1L and CLIP-170) were accompanied by an increase in the levels of the 10 and 8 kDa carboxyl-terminal fragments of APP (not shown). These fragments have been shown to be generated by ß- and 
-secretases and are direct precursors of Aß and p3, respectively. This result indicated that the interaction of full-length PS-1 with full-length CLIP-70 is prerequisite of a functional 
-secretase and that its disruption has consequences on Aß production.
4. Disruption of the PS1/CLIP-170 complex is associated with a decrease in Aß uptake: a possible link between PS1 and Aß uptake through ApoE/LRP
As CLIP-170 was originally isolated as a protein capable of conjoining endosomes to the microtubules (MTs), we explored the possibility that its interaction with PS1 may be involved, not only in the generation but, the internalization of extracellular Aß. A tighter binding of the two proteins (induced by either FAD-linked mutations or high intracellular Ca2+) may be linked to a deficiency in endocytosis/lysosomal degradation, a process known to be affected early in AD. Aß can be internalized and cleared from the media efficiently when preincubated with lipoprotein particles containing apoE and subsequent binding to the lipoprotein receptor-related protein (LRP) in neurons. SY5Y cells, known to express LRP, are capable of its uptake (Fig. 2C
), which is inhibited by extracellular receptor-associated protein (RAP) (Fig. 2B
). Conversely, apoE is taken up by cells more readily when decorated with lipids and Aß. After internalization, Aß itself is either degraded or transported through the cell, a process perhaps both cell and apoE isoform specific.
|
To test whether PS1 alone and its interaction with CLIP-170 is important to Aß uptake by SY5Y cells, we examined the effect of the transfection with PS1 antisense oligonucleotides (Fig. 2B
). These treatments reduce the expression of PS1 by 42% (Fig. 2A
). Next, we examined the effect of expression of the BDPs (CLIP-C, PS1L) on Aß uptake. Both treatments reduce the amount of Aß taken up by the cells (Fig. 2B
). Evidently, PS1 has a primary role in mechanisms of Aß secretion as well as uptake from cells, and both roles are mediated through PS1’s interaction with CLIP-170. This reflects an apparent complication in a strategy aiming to target the interaction of the two proteins as a putative AD therapy.
5. APP and PS1 colocalize to the microtubules through a specific subcellular fraction: absence of Notch-IR in that fraction suggests distinct pathways for PS1-dependent proteolytic processing of APP and Notch, respectively
Because our findings imply an obligatory association of PS1 with the cytoskeleton, we used immunofluorescent confocal microscopy to determine whether PS1 and APP colocalize to MTs in SY5Y and D9-12 cells. Indeed, PS1 and APP fluorescence colocalized with tubulin (not shown). We prepared MTs from the cytoplasmic fraction of HeLa or SY5Y cell homogenates by polymerizing tubulin with taxol, followed by low-speed centrifugation. We then added to the polymerization reaction distinct subcellular fractions from the same cell extracts known to contain APP, PS1, and Notch-like immunoreactivity and examined whether they could be cosedimented with the MTs. Subcellular fractions containing carboxyl-terminal fragments of APP and full-length PS1 were cosedimented quantitatively with the MTs in the absence but not the presence of ATP/NaCl (not shown). Inclusion of purified recombinant GST-BDPs in the polymerization reaction prohibited the coprecipitation of vesicles immunoreactive for APP and PS1 with the MTs (not shown). Even though Notch immunoreactivity could be readily detected in the homogenate at low speed and at lower levels in the high-speed pellet of SY5Y cells transfected with Tan-1 and cyt-Tan-1 cDNA, Notch-like immunoreactivity could not be cosedimented with the MT pellet (not shown).
CONCLUSIONS AND SIGNIFICANCE
Figure 3
represents our working model on the role of presenilin in Aß-related pathways and how this may be mediated through its interaction with a cytoskeletal protein, namely, CLIP-170. It is speculated that CLIP-170 is required for the initial docking of the vesicles loaded with APP metabolites and tagged with PS1. Subsequent retrograde transport of these vesicles toward the (–) end of the microtubules, believed to be orchestrated by dynein and dynein-associated protein complexes can occur between the Golgi compartment and the ER or between the plasma membrane and cytoplasmic degradative machinery (endosomal/lysosomal compartments). Thus, a possible involvement/triggering of the PS1/CLIP-170 interaction in/of retrograde transport within cells could be associated with either the formation (Golgi/ER and at plasma membrane) or degradation of Aß. If so, it would be interesting to determine the contribution of each of these possible pathways toward the buildup of Aß both intracellularly and extracellularly. The increase in the amyloid load associated with pedigree members carrying FAD-linked mutations in presenilins could be attributed not only to an increase in the production/secretion of certain Aß species, but to a decrease in their clearance. As Aß uptake is intricately linked to ApoE and LRP, our model integrates the major susceptibility gene products in AD—APP, PS1, and ApoE—providing a firm basis for testable hypotheses and new drug discovery. The specificity and physiological significance of the above interactions of PS1 with CLIP-170 for AD-related pathways is being explored further, including the generation of CLIP-170 knockout animals.
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0980fje; doi: 10.1096/fj.02-0980fje 
This article has been cited by other articles:
|
|
 |
|
  G. B. Stokin, A. Almenar-Queralt, S. Gunawardena, E. M. Rodrigues, T. Falzone, J. Kim, C. Lillo, S. L. Mount, E. A. Roberts, E. McGowan, et al. Amyloid precursor protein-induced axonopathies are independent of amyloid-{beta} peptides Hum. Mol. Genet., November 15, 2008; 17(22): 3474 - 3486. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Mahoney, A. L. Parks, D. A. Ruddy, S. Y. K. Tiong, H. Esengil, A. C. Phan, P. Philandrinos, C. G. Winter, R. Chatterjee, K. Huppert, et al. Presenilin-Based Genetic Screens in Drosophila melanogaster Identify Novel Notch Pathway Modifiers Genetics, April 1, 2006; 172(4): 2309 - 2324. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. C. FEWLASS, K. NOBOA, F. X. PI-SUNYER, J. M. JOHNSTON, S. D. YAN, and N. TEZAPSIDIS Obesity-related leptin regulates Alzheimer's A{beta} FASEB J, December 1, 2004; 18(15): 1870 - 1878. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-g. Lee, P. I. Moreira, X. Zhu, M. A. Smith, and G. Perry Staying Connected: Synapses in Alzheimer Disease Am. J. Pathol., November 1, 2004; 165(5): 1461 - 1464. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
